Pdf Figure Extractor
v1.0.0从PDF论文中精确提取Figure图片，自动分析PDF结构、定位caption位置、裁剪干净图形，并验证图片质量。支持学术新闻稿、论文写作等场景的自动化图片处理。
⭐ 0· 181·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (PDF figure extraction) matches the SKILL.md: it contains concrete Python examples that analyze PDF text blocks, locate captions, compute crop rectangles, render, and save images. All requested operations are consistent with extracting figures.
Instruction Scope
Instructions operate only on a PDF input (pdf_path, page_num) and explain how to find captions and crop images. They do not instruct reading unrelated filesystem paths or environment secrets. Note: the instructions rely on heuristics and hardcoded coordinate templates which may need tuning for diverse PDFs.
Install Mechanism
This is an instruction-only skill with no install spec. The example code uses the 'fitz' module (PyMuPDF) but the skill does not declare that dependency or advise installing it. That omission reduces completeness/coherence but is not malicious.
Credentials
No environment variables, credentials, or config paths are requested. The skill only requires access to PDFs provided by the user, which is proportionate to its purpose.
Persistence & Privilege
always is false and there is no indication the skill requests persistent privileges or modifies system/agent-wide settings. Autonomous invocation is allowed (platform default) but not excessive here.
Assessment
This skill is an instruction-only recipe for extracting figures from PDFs and appears to do what it claims. Before using it: (1) ensure the agent/runtime has Python and the PyMuPDF package (import fitz) installed — the SKILL.md uses fitz but does not list this dependency; (2) run it on non-sensitive sample PDFs first to tune crop coordinates and verify captions are removed; (3) process private or confidential PDFs in a trusted, isolated environment since the skill will read full PDF contents; (4) if you need automated installs, add a controlled install step (e.g., pip install PyMuPDF) rather than relying on implicit environment assumptions. If you want higher assurance, ask the author for an explicit dependency list and a test vector.Like a lobster shell, security has layers — review code before you run it.
latestvk9772tyjvh59r3z846nrv9ptqh834smr
License
MIT-0
Free to use, modify, and redistribute. No attribution required.